Databricks Machine Learning Associate exam dumps

Databricks Machine Learning Associate practice question 628 of 656

Databricks Certified Machine Learning Associate. Associate level, Databricks. Free question with the correct answer and a full explanation.

Databricks Machine Learning Associate Question 628

Select 1

You are tasked with deploying a machine learning model in a Databricks environment to predict customer churn. The model predictions need to be consumed by a customer support dashboard in near real-time, but historical data analysis also needs to be conducted weekly to identify trends. Which combination of model serving approaches would best meet these requirements?

  1. A

    Batch serving for the dashboard and real-time serving for historical analysis

  2. B

    Real-time serving for the dashboard and batch serving for historical analysis

  3. C

    Streaming serving for both the dashboard and historical analysis

  4. D

    Real-time serving for the dashboard and streaming serving for historical analysis

  5. E

    Batch serving for both the dashboard and historical analysis

Show answer and explanation

Correct answer: B

Explanation

The correct solution is to use real-time serving for the customer support dashboard to ensure timely predictions and batch serving for historical data analysis to process and analyze large data volumes efficiently. This combination balances latency, scalability, and complexity for the given requirements.

  • A. Incorrect.

    Batch serving for the dashboard would introduce latency and is not suitable for near real-time needs. Real-time serving for historical analysis is unnecessary, as historical trends are typically computed in batch mode.

  • B. Correct.

    Real-time serving ensures near real-time predictions for the customer support dashboard, while batch serving can efficiently process large volumes of historical data periodically for trend analysis.

  • C. Incorrect.

    Streaming serving is more suited for continuous data ingestion and processing in use cases like IoT or event-driven systems, but it adds unnecessary complexity here. Additionally, historical analysis is best handled in batch mode.

  • D. Incorrect.

    Streaming serving might work for the dashboard but is not ideal for historical data processing, which is better suited for batch serving. This approach adds unnecessary overhead in this scenario.

  • E. Incorrect.

    Batch serving would be too slow for the dashboard's near real-time requirements and would not meet the performance needs of the given scenario.

Timed practice exam

Take a Databricks Machine Learning Associate practice test under exam conditions

48 questions in 90 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam

More Databricks Machine Learning Associate practice questions

  1. 629.A retail company wants to implement a machine learning model to recommend products to users based...
  2. 630.A company wants to deploy a machine learning model to recommend personalized products to users....
  3. 631.You are tasked with deploying a custom machine learning model to a Databricks model serving...
  4. 632.You are tasked with deploying a custom machine learning model to a Databricks model serving...
  5. 633.You have trained a custom machine learning model and want to deploy it to a model endpoint in...
Back to the Databricks Machine Learning Associate exam dumps